Distributed Denial of Service (DDoS) attacks are among the primary concerns in internet security today. Machine learning can be exploited to detect such attacks. In this paper, a multi-layer perceptron model is proposed and implemented using deep machine learning to distinguish between malicious and normal traffic based on their behavioral patterns. The proposed model is trained and tested using the CICDDoS2019 dataset. To remove irrelevant and redundant data from the dataset and increase learning accuracy, feature selection is used to select and extract the most effective features that allow us to detect these attacks. Moreover, we use the grid search algorithm to acquire optimum values of the model’s hyperparameters among the parameters’ space. In addition, the sensitivity of accuracy of the model to variations of an input parameter is analyzed. Finally, the effectiveness of the presented model is validated in comparison with some state-of-the-art works.

Classical methods are not usually efficient, to solving nonlinear control problems and especially Nonlinear Distributed parameter systems Optimal Control Problems (NOCP). In this paper we introduce a new approach for solving this class of problems by using Non Linear Programming Problem (NLPP). First, we transfer the original problem to a new problem in form of calculus of variations. The next step we discrete the new problem and solve it by using NLPP packages. Moreover, a NLPP is transferred to a Linear Programming Problem (LPP) which empowers us to use powerful LP software. Finally, efficiency of our approach is confirmed by some numerical examples.

parameter estimation is one of the main tasks in the hydrologic modeling. parameter values can either be specified logically or blindly calibrated. In Distributed physics-based models, it is arguably possible to specify parameters using catchment characteristics, hydrologic knowledge, the physics behind the parameters and how they function in the model. Such logic-based parameter specification is called as parameter allocation. This study tries to practice this modeling approach using the MIKE SHE model and simulate the overland flow in the Ziarat watershed. The model was executed using both approaches for a certain period from 01. 23. 2013 to 09. 21. 2014, and for the period from 04. 15. 2016 to 09. 21. 2017 as validation. The results of the simulation based on each of the approaches were evaluated by the Nash-Sutcliffe and Kling-Gupta efficiency criteria. Based on these efficiency criteria, the model in the parameter allocation approach has a good performance, and shows consistency in the validation period. Regarding the water balance, the results of the allocation approach are more resoanable and meaningful. Based on this, it can be concluded that spending more time to better understand the watershed charactristics and parameters of the model leads to more acceptable and consistent results that reduces the need for calibration significantly.

In this paper, a Multi-Input Multi-Output (MIMO) model Reference Adaptive Control (MRAC) scheme for a flexible rotating arm is developed. In order to construct a reference model to be followed by this Distributed parameter system, a finite element method is used to approximate the behavior of the arm. An input error direct adaptive control algorithm is utilized as the control approach to account for parameter uncertainty. Assuming the same approximation and structure as the model for the actual system, the stability analysis of the proposed controller will be straightforward. Simulation results are provided to illustrate the performance of the proposed algorithm in the presence of disturbance and uncertainties. Also, the proposed algorithm results are compared with those of a conventional PD controller.

In this paper a new learning automata-based algorithm is proposed for learning of parameters of a Bayesian network. For this purpose, a new team of learning automata which is called eDLA is used. In this paper the structure of Bayesian network is assumed to be fixed. New arriving sample plays role of the random environment and the accuracy of the current parameters generates the random environment reinforcement signal. Linear algorithm is used to update the action selection probability of the automata. Another key issue in Bayesian networks is parameter learning under circumstances that new samples are incomplete. It is shown that new proposed method can be used in this situation. The experiments show that the accuracy of the proposed automata based algorithm is the same as the traditional enumerative methods such as EM. In addition to the online learning characteristics, the proposed algorithm is in accordance with the conditions in which the data are incomplete and due to the use of learning automaton, has a little computational overhead.

Distributed sediment yield models may be improved by using remote sensing to identify land cover classes and RADAR precipitation data as input. Sediment yields are calculated separately for each land cover class and then routed downstream. Calibration and validation are accomplished with data from an experimental watershed. The model simulated run off hydrographs which correlated closely with those observed. The sediment discharge graphs were comparable to observed data.